This paper uses Roy's model of sorting behavior to study welfare implication of current health care data production infrastructure that relies on solicitation of research subjects. We show that due to severe adverse-selection issues, directionality of bias cannot be established and welfare may...

I demonstrate that to achieve dynamic efficiency, the optimal share of total surplus that a social payer should transfer to an innovating industry for a current asset depends on the marginal product of investment and the share of profits invested by the industry on the current asset and not on...

This paper builds on the methods of local instrumental variables developed by Heckman and Vytlacil (1999, 2001, 2005) to estimate person-centered treatment (PeT) effects that are conditioned on the person's observed characteristics and averaged over the potential conditional distribution of...

The United States aspires to use information from comparative effectiveness research (CER) to reduce waste and contain costs without instituting a formal rationing mechanism or compromising patient or physician autonomy with regard to treatment choices. With such ambitious goals, traditional...

There is growing concern that the increasing use of machine learning and artificial intelligence-based systems may exacerbate health disparities through discrimination. We provide a hierarchical definition of discrimination consisting of algorithmic discrimination arising from predictive scores...

This paper examines the impact of California's hospital closures occurring from 1995-2011 on adjusted inpatient mortality for time-sensitive conditions: sepsis, stroke, asthma/chronic obstructive pulmonary disease (COPD) and acute myocardial infarction (AMI). Using a difference- in-difference...

Public technology assessments in general and Comparative Effectiveness Research (CER) in particular have been justified by offsetting benefits of improving patient health and reducing health care spending. However, little conceptual and empirical understanding exists concerning the quantitative...

Features of Part D gave rise to broad concern that the drug benefit would negatively impact prescription utilization among the six million dual eligible beneficiaries, either during the transition from state Medicaid to Part D coverage, or in the long-run. At the same time, Part D contained...

Under the assumption of no unmeasured confounders, a large literature exists on methods that can be used to estimating average treatment effects (ATE) from observational data and that spans regression models, propensity score adjustments using stratification, weighting or regression and even the...